CN101000603A - Patent classification method - Google Patents
Patent classification method Download PDFInfo
- Publication number
- CN101000603A CN101000603A CNA2006101488323A CN200610148832A CN101000603A CN 101000603 A CN101000603 A CN 101000603A CN A2006101488323 A CNA2006101488323 A CN A2006101488323A CN 200610148832 A CN200610148832 A CN 200610148832A CN 101000603 A CN101000603 A CN 101000603A
- Authority
- CN
- China
- Prior art keywords
- speech
- classification
- specification digest
- word
- patent classification
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Landscapes
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
A method for classifying patent includes obtaining a patent abstract, carrying out judgment word treatment on said abstract, filtering word with no relation to topic in said abstract out, carrying out stem normalization treatment on filtered word, calculating weight of each word, classifying said patent and demonstrating classification result of patent.
Description
Technical field
The present invention relates to a kind of sorting technique, especially relate to a kind of patent classification method
Background technology
Frequent along with science and technology development and business activity, protection of Intellectual Property Rights is more and more paid attention to by people, and patent is exactly one of most important means in the intellecture property.Because most of new technologies all are the form appearance with patent when starting to walk, thereby have stored a large amount of technical information in patent database.Utilize these patent databases can understand the state-of-the-art technology of every profession and trade, avoid overlapping development and generation tortious, can analyze rival's technological development situation and strategy, can also analyze the development of whole industry etc.Nowadays, all open its patent database in world many countries and area, as United States Patent (USP) trademark office (United States Patent and Trademark Office) patent database, EUROPEAN PATENT OFFICE (European Patent Office) patent database, China national Department of Intellectual Property (State Intellectual Property Office of China) patent database etc.
Yet patent research is the time-consuming work of effort again, because not all patent all has researching value.How obtaining the patent information useful to company from numerous numerous and jumbled patents, how the patent information that finds is made further statistics and technical Analysis, is a great problem in the patent research.So a kind of patent classification system and method need be provided, and it utilizes patent content with patent automatic classifying, and can show the patent classification result.
Summary of the invention
Fundamental purpose of the present invention is to provide a kind of patent classification method, and it can finish the classification of patent automatically.
In order to achieve the above object, said method comprising the steps of:
(1) obtaining of patent is used to obtain the specification digest of one piece of patent;
(2) the disconnected speech of patent is used for the described specification digest speech that breaks is handled;
(3) filtering module of irrelevant speech, the speech that is used for haveing nothing to do with theme in the described specification digest filters;
(4) normalization of stem is used for that the speech after filtering is carried out stem normalization and handles;
(5) calculating of weight is used to calculate the weight of each speech;
(6) classification of patent is used for described patent classification;
(7) displaying of patent is used to show the patent classification result.
Utilize the present invention, can finish the classification of patent automatically, and show the patent classification result.
Description of drawings
Fig. 1 is hardware structure figure of the present invention.
Fig. 2 is an operation process chart of the present invention.
Embodiment
The present invention is described in detail below in conjunction with accompanying drawing.
As shown in Figure 1, hardware configuration of the present invention comprises: application server 1, client computer 2, database 3, described client computer 2 links to each other with the described server 1 of application by network, network can be an intranet (Intranet), also can be internet (Internet) or other type communication network, described application server 1 links to each other with database 3 by data line, and application server 1 is used for patent is classified.Patent in the patent of present embodiment refers in a certain theme, described client computer 2 is used to show the patent classification result, described database 3 is used to store patent information and patent classification result.Above-mentioned patent information refers to the full detail of the patent of open or bulletin, comprises claims, instructions, and, specification digest, Figure of abstract, applicant, the applying date, patentee etc.
As shown in Figure 2, the invention provides a kind of patent classification method, this method may further comprise the steps:
(1) obtaining of patent is used to obtain the specification digest of one piece of patent, and the user realizes on client computer 2;
(2) the disconnected speech of patent is used for the described specification digest speech that breaks is handled, and promptly according to space or punctuation mark sentence is divided into speech, realizes by application server 1 accessing database 3;
(3) filtering module of irrelevant speech, the speech that is used for haveing nothing to do with theme in the described specification digest filters, and realizes by application server 1 accessing database 3;
(4) normalization of stem is used for that the speech after filtering is carried out stem normalization and handles, and the different shape that is about to same speech is normalized to same form, realizes by application server 1 accessing database 3;
(5) calculating of weight is used for according to each speech calculating the weight of this speech in this patent in the frequency that this specification digest occurs, and realizes by application server 1 accessing database 3;
(6) classification of patent is used for the result of described patent classification according to weight calculation classified above-mentioned patent;
(7) displaying of patent is used for the patent classification result is illustrated on the client computer 2.
Claims (1)
1. patent classification method is characterized in that said method comprising the steps of:
(1) obtaining of patent is used to obtain the specification digest of one piece of patent;
(2) the disconnected speech of patent is used for the described specification digest speech that breaks is handled;
(3) filtering module of irrelevant speech, the speech that is used for haveing nothing to do with theme in the described specification digest filters;
(4) normalization of stem is used for that the speech after filtering is carried out stem normalization and handles;
(5) calculating of weight is used to calculate the weight of each speech;
(6) classification of patent is used for described patent classification;
(7) displaying of patent is used to show the patent classification result.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CNA2006101488323A CN101000603A (en) | 2006-12-29 | 2006-12-29 | Patent classification method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CNA2006101488323A CN101000603A (en) | 2006-12-29 | 2006-12-29 | Patent classification method |
Publications (1)
Publication Number | Publication Date |
---|---|
CN101000603A true CN101000603A (en) | 2007-07-18 |
Family
ID=38692580
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CNA2006101488323A Pending CN101000603A (en) | 2006-12-29 | 2006-12-29 | Patent classification method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN101000603A (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2008128445A1 (en) * | 2007-04-23 | 2008-10-30 | Huawei Technologies Co., Ltd. | Method and system for content classification |
CN111858941A (en) * | 2020-07-28 | 2020-10-30 | 中译语通科技股份有限公司 | Patent classification method and device |
-
2006
- 2006-12-29 CN CNA2006101488323A patent/CN101000603A/en active Pending
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2008128445A1 (en) * | 2007-04-23 | 2008-10-30 | Huawei Technologies Co., Ltd. | Method and system for content classification |
US8286240B2 (en) | 2007-04-23 | 2012-10-09 | Huawei Technologies Co., Ltd. | Method and system for content categorization |
US8510832B2 (en) | 2007-04-23 | 2013-08-13 | Huawei Technologies Co., Ltd. | Method and system for content categorization |
CN111858941A (en) * | 2020-07-28 | 2020-10-30 | 中译语通科技股份有限公司 | Patent classification method and device |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN105138652B (en) | A kind of enterprise's incidence relation recognition methods and system | |
WO2006015054A3 (en) | Methods, systems and computer program products for performing subsequent transactions for prior purchases | |
CN104899268A (en) | Distributed enterprise information vertical searching method | |
WO2003107124A3 (en) | Computerized system and method of performing insurability analysis | |
WO2005119547A3 (en) | System and method for organizing price modeling data using hierarchically organized portfolios | |
CN101044472A (en) | Methods and systems for semantic identification in data systems | |
CN102346901A (en) | Internet medicine trading subject credit assessment system and method | |
CN102073641A (en) | Method, device and program for processing consumer-generated media information | |
Cardon et al. | Two Paths of Glory—Structural Positions and Trajectories of Websites within Their Topical Territory | |
CN110968571A (en) | Big data analysis and processing platform for financial information service | |
CN112948391A (en) | Performance assessment method, system, terminal and readable storage medium | |
CN108985707A (en) | A kind of method of quick judgement resume content authenticity | |
CN201114128Y (en) | Enterprise search engine device | |
CN104731908A (en) | ETL-based data cleaning method | |
CN101000603A (en) | Patent classification method | |
JP2009266039A5 (en) | ||
CN104699753A (en) | Intellectual property inquiry system based on cloud database | |
EP1489810A3 (en) | System and method for providing security mechanisms for data warehousing and analysis | |
CN201936346U (en) | IT audit accessory system | |
Joyfong et al. | Preparation of Smart Card Data for Food Purchase Analysis of Students through Process Mining | |
CN106204252A (en) | Internal credit and debt remaining sum identification, the method and system gathering and checking | |
CN110533392A (en) | A kind of realization method and system confirming capital settlement attribution data unit | |
CN102722496A (en) | Patent classification method | |
CN102270242B (en) | Computer-aided corpus extraction method | |
CN101000613A (en) | IPC number technology subject inquiry method |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C02 | Deemed withdrawal of patent application after publication (patent law 2001) | ||
WD01 | Invention patent application deemed withdrawn after publication |
Open date: 20070718 |